UMEA UNIVERSITYDepartment of Physics

Lab PMApril/May 2013

Direct absorption and wavelength modulationspectroscopyDetection of CO2

Aim: To compare the performance of direct absorption spec-troscopy and wavelength modulation spectroscopy for de-tection of CO2.

Prerequisites: Intensive study of these lab instructions. All questionsmarked with ”Question” must be answered before comingto the lab.

Lab instructions for the course: Laser-based spectroscopic techniques

aleksandra.foltynowicz@physics.umu.se

jonas.westberg@physics.umu.se

Department of PhysicsUmea UniversitySE-901 87 Umea, Sweden

1 Introduction

Direct absorption spectroscopy (DAS) is the simplest realization of laser-based absorption tech-niques. In DAS, a beam of a tunable laser is sent through a gas sample and the transmittedintensity is measured with a detector. If the frequency of the light is close to an atomic ormolecular transition the light is absorbed and the transmitted intensity decreases. The con-centration of the absorbing species can be calculated from the relative change of the intensityaccording to Lambert-Beer’s law. The drawback of this simple technique is that its sensitivity islimited by low frequency noise in the signal, which originates mainly from laser intensity noise,mechanical instabilities and other external fluctuations. This noise is often referred to as the1/f -noise, because its power spectrum scales roughly with inverse frequency. The influence ofthe 1/f -noise can be greatly reduced by shifting the detection to higher frequencies with theuse of modulation techniques. The most common of these techniques is wavelength modulationspectroscopy (WMS). In WMS, the frequency of the laser light is modulated at a high frequencyf (usually in the kHz range), which leads to a modulation of the transmitted intensity in thevicinity of the transition. The resulting periodic detector signal is then demodulated at a mul-tiple integer, nf , of the modulation frequency, where n = 1, 2, 3, ..., with the use of a lock-inamplifier. This procedure significantly increases the sensitivity; while DAS typically can detectabsorbances in the 10−2 to 10−3 range, the WMS technique can detect absorbance down to10−5 to 10−6.

There are other laser-based spectroscopic techniques that can provide detection limits equiv-alent to or superseding those of WMS. However, these techniques are not as cost-effective, easyto implement and robust as the WMS technique. As a result, WMS is one of the most widelyused spectroscopic techniques for industrial applications today.

In this laboratory exercise you will familiarize yourself with experimental implementation ofthe DAS and WMS techniques, and compare their performance (sensitivity) using CO2 as apilot species.

2 Direct absorption spectroscopy

A schematic experimental setup of DAS is shown in Figure 1. A laser beam with power P0

is sent through an absorbing sample of length L (cm) and the transmitted power P (ν) (W)is measured with a photodetector as a function of optical frequency ν. The wavelength of thelaser is tuned by applying an external voltage signal to the laser driver. The detector signal isproportional to the incident power

SD(ν) = GηP (ν), (1)

where η is the detector responsivity (A/W) and G is the detector amplifier’s gain (V/A).The transmitted power is related to the incident power through the Beer-Lambert’s law,

which states that

P (ν) = P0e−α(ν)L, (2)

where ν is the frequency of the light (Hz) and α(ν) is the absorption coefficient (cm−1), whichis given by

α(ν) = σ(ν)(N1 −

g1g2N2

), (3)

3

Laser

Functiongenerator

α(υ)detector

S (υ)D

I(υ)I0

sweep L

Figure 1: The basic principle of direct absorption spectroscopy. As light passes through amedium containing an absorber it is attenuated according to Beer-Lambert’s law. Thetransmitted light is focused on a detector which produces a signal that is proportionalto the incident power. By sweeping the laser frequency, with the aid of the functiongenerator, it is possible to obtain an absorption spectrum of the gas.

where σ(ν) is the absorption cross-section of the transition (cm2), N1 and N2 are the den-sities of absorbers in the lower and upper energy levels of the transition, and g1 and g2 arethe degeneracies of the two levels, respectively. In near- and mid-infrared spectroscopy, theabsorption coefficient is often expressed in terms of an integrated molecular linestrength, S[cm−1/(molecule·cm−2)], as

α(ν) = Snxχ(ν), (4)

where nx is the density of absorbers (cm−3) and χ(ν) is the frequency dependent area-normalizedlineshape function (1/cm−1), which has the following normalization requirement

∫ ∞0

χ(ν) dν = 1. (5)

The integrated molecular linestrength, S, is tabulated in databases, e.g. HITRAN, and it takesinto account population distribution at a given temperature and energy level degeneracies.

It is convenient to rewrite equation (4) in terms of the total pressure of the gas, p (atm), andconcentration of the absorbers, crel (dimensionless), as

α(ν) = S′crelpχ(ν), (6)

where S′ is the integrated gas linestrength (cm−2/atm), related to the integrated molecularlinestrength, S, through

S′ = 2.6868 · 1019T0TS, (7)

where T is the temperature of the gas (K) and T0 is the reference temperature equal to 273.15 K.The constant in equation (7) is the Loschmidt number, which represents the density of moleculesof an ideal gas at 0 C and 1 atm.

The exponent in the Beer-Lambert law, equation (2), is called the absorbance and it isproportional to the concentration of the absorbing species. It can be calculated from the ratioof the transmitted and incident power, i.e. as,

α(ν)L = lnP0

P (ν). (8)

4

For an optically thin medium, for which α(ν)L 1, equation (2) can be Taylor expanded

P (ν) = P0

[1− α(ν)L

], (9)

and the absorption of the analyte is directly proportional to the relative change of power,

α(ν)L =P0 − P (ν)

P0=

∆P (ν)

P0. (10)

The maximum value of the absorbance for a given transition occurs at resonance. Thereforeit is convenient to introduce the on-resonance absorbance, α0L, as

α0L = S′crelpχ0L, (11)

where χ0 is the peak-value of the area-normalized lineshape function, i.e. χ0 = χ(ν = ν0),where ν0 is the center frequency of the transition.

2.1 Absorption lineshapes

Earlier we introduced the lineshape function, χ(ν), although its physical origin was not statedclearly. The reason for introducing this entity is that absorption does not occur at a singlewavelength. Instead, the absorption lines are broadened due to the limited lifetime of the energylevels involved in the transition and due to the thermal motion of the molecules. The lineshapefunction describes the relative absorption around the center frequency of the transition. Thereare a number of different mechanisms that affect the broadening of the lineshapes. Here, wewill focus on two of these that are dominating for the conditions that you will encounter in thislab; the Doppler broadening and the collision (pressure) broadening.

2.1.1 Doppler broadening

The constant motion in random directions of the absorbing molecules in a gaseous medium leadsto a distribution of velocities which can be described by the Maxwell-Boltzmann distributionfunction, given by

fG(vz) =1√πue−v

2z/u

2, (12)

where vz is the molecular velocity in the z-direction (m/s), and u is the most probable velocity ofa thermal distribution of molecules (m/s) given by

√2kBT/m, where kB is Boltzmann’s constant

(J/K), T the temperature (K) and m the molecular mass (kg). Whenever the molecules havea velocity-component in the direction of propagation of the incoming light their absorptionfrequency is slightly shifted due to the Doppler effect. Such molecules do not absorb at afrequency ν0 but rather at the shifted frequency ν0(1 ± vz/c). This means that light with acertain frequency can only interact with a specific thermal group of molecules which leads toa distribution of absorption frequencies and subsequently an inhomogeneous broadening of thetransition. The lineshape function describing this phenomenon has a Gausssian form and canbe written as

χG(νd) =

√4 ln 2

π

c

δνDexp

[−4 ln 2

ν2dδν2D

], (13)

5

where νd is the laser frequency detuning from the molecular resonance given by νd = ν − ν0, cthe speed of light (cm/s), and δνD is the Doppler width (full-width-half-maximum, FWHM) ofthe Gaussian profile (Hz) given by,

δνD = 2ν0

√2 ln 2kBT

mc2. (14)

This can be rewritten as

δνD = 7.162 · 10−7ν0

√T

M, (15)

where M is the molecular weight (g/mol).

Question What is the Doppler width (FWHM) of a CO2 transition at 1570.3 nm and at room-temperature?

δνD = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.1.2 Collision (pressure) broadening

If the absorbing molecules suffer from frequent collisions with other molecules the life timeof the molecular states is shortened. This occurs predominantly at higher pressures wherethe molecular density is larger and more collisions are likely to occur. This causes homogenousbroadening of the absorption profile (it affects all molecules equally) and gives rise to a lineshapeof Lorentzian form, which is given by

χL(νd) =c

π

δνL/2

(νd)2 + (δνL/2)2, (16)

where νd is again the frequency detuning and δνL is the collision broadened full-width-half-maximum (FWHM) linewidth (Hz), given by

δνL = 2∑i

γipi (17)

where γi is the half-width-half-maximum pressure broadening coefficient (Hz/atm) for collisionswith a species i and pi is its partial pressure (atm). Note that in general the homogeneous broad-ening is given by a sum of contributions of different broadening mechanisms, including collisionbroadening, natural broadening, saturation broadening and transit time broadening. However,at high pressure conditions the collision broadening is so large that the other contributions canbe neglected.

Question What is the pressure broadened linewidth (FWHM) of a CO2 transition with a self-pressure broadening coefficient of 2.7 GHz/atm in a pure CO2 gas (crel = 1) at a pressure of600 Torr?

δνL = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Figure 2 shows the Doppler lineshape and the Lorentzian lineshape with equal linewidths.

6

0

Detuning frequency, νd

Absorption

lineshap

e

χG(νd)χL(νd)

Figure 2: The Doppler and Lorentzian absorption lineshapes.

Question Which broadening mechanism dominates for this particular CO2 transition (1570.3nm) and experimental configuration (600 Torr of pure CO2 gas and room temperature)?

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Question Calculate the on-resonance absorbance α0L for pure CO2 detected at a transitionclose to 1570.3 nm for the dominating broadening mechanism at a pressure of 600 Torr and roomtemperature. The integrated gas linestrength can be found in Figure 3.

α0L = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1568 1569 1570 1571 1572 1573 1574

−5

0

5

10

15

20

x 10−24

Wavelength [nm]

S[cm

−1/m

olecule

cm−2]

Figure 3: Integrated molecular linestrengths for transitions in CO2 in the wavelength region1567 nm to 1575 nm from the HITRAN database.

2.2 Detection limit

There are several ways to determine the limit of detection (LOD) of a spectroscopic system. Oneway is to define the detection limit, (α0L)LOD, as the ratio of the on-resonance absorbance to

7

the signal-to-noise ratio (SNR) of the detector signal measured for that particular absorbance,viz. as

(α0L)LOD =α0L

SNR. (18)

The SNR can be estimated by comparing the peak value of the detector signal, SD(ν = ν0),evaluated on resonance for a given α0L, to three times the standard deviation of the noise inthe background signal, σbg,

SNR =SD(ν = ν0)

3σbg. (19)

Question Calculate the SNR for a system with an on-resonance absorption signal of 1 V anda standard deviation of the noise in the background signal of 16.7 mV. What is the minimumdetectable absorbance if the on-resonance absorbance in this measurement is 2× 10−2?

SNR ≈ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

(α0L)LOD ≈ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.2.1 Shot noise limit

The fundamental noise limit in direct absorption spectroscopy is the shot noise limit, whichoriginates from the quantum nature of light, namely that the photons impinging on the detectorhave a Poissonian distribution. Since this is the fundamental noise limit, a system that is saidto be shot noise limited shows its lowest attainable detection limit, which is given by

(α0L)shotLOD =

√2eB

NηP0, (20)

where e is the electron charge (C), B is the detection bandwidth (Hz), N is the number ofaverages, η is the responsivity of the detector (A/W) and P0 is the power incident on thedetector (W). In reality this limit is seldom reached with DAS due to the existence of other,more dominant, types of noise.

Question What is the shot noise limit (without averaging) for a DAS system in which 1 mWof power is incident on a detector with responsivity of 1 (A/W) and the electronic bandwidthis set to 1 Hz?

(α0L)shotLOD: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

8

3 Wavelength modulation spectroscopy

A schematic experimental setup of WMS is shown in Figure 4. The wavelength of the laseris swept across the transition at a low frequency. In addition, the lock-in amplifier provides asinusoidal modulation at a much higher frequency, which is combined with the sweep and fedto the external modulation input of the laser driver. The detector signal, SD, is demodulatedby the lock-in amplifier at frequency nf , which yields a WMS signal, Sn(ν).

Laser

Functiongenerator

Lock-inamplifier

f

S (υ)n

α(υ)detector

S (υ)D

I(υ)I0

+f

sweepL

Figure 4: Basic experimental setup of the WMS technique.

In this lab you will be using a diode laser, which is very useful for WMS because its wavelengthcan be easily modulated by applying a sinusoidal modulation to the injection current, accordingto

ic(t) = ic + ia cos(2πft), (21)

where ic is the center injection current, ia the injection current modulation amplitude, f isthe modulation frequency and t is the time. This leads to a corresponding modulation of thewavelength. Although the name of the technique suggests that it is the wavelength that ismodulated it has been found more convenient to express the modulation in terms of frequencyunits, i.e. as

ν(t) = νc + νa cos(2πft), (22)

where νc is the center frequency of the laser and νa the frequency modulation amplitude.According to equations (1) and (9) the detector signal for optically thin samples is given by

SD[ν(t)] = GηP0 −GηP0α[ν(t)]L. (23)

The second term, which is proportional to absorbance, is periodically modulated due to themodulation of the optical frequency. It can be expressed in terms of a Fourier series as

SD[ν(t)] =∞∑n=0

Sn(ν) cos(2πnft), (24)

where Sn(ν) is the nth Fourier coefficient of the detector signal, given by

9

Sn(ν) =2− δn0τ

∫ τ

0SD[ν(t)] cos(2πnft) dt, (25)

where δn0 is the Kronecker’s delta, which is equal to 1 if n = 0 and 0 if n 6= 0. The integrationtime τ must be equal to a multiple of the inverse of the modulation frequency, 1/f . Usingequation (23) for the detector signal and equation (6) for α(ν), the nth Fourier coefficient canbe written as

Sn(ν) = GηP0S′crelpL

2

τ

∫ τ

0χ(t) cos(2πnft) dt, (26)

There exist analytical solutions to the integrals in equation (26) for the Lorentzian andGaussian lineshape functions. The first six (n = 1, 2, ..., 6) Fourier coefficients of modulatedLorentzian lineshape functions are plotted in Figure (5). For other lineshapes the integrals canbe evaluated numerically. The nth Fourier coefficient of the detector signal can be measuredwith the use of a lock-in amplifier. The lock-in amplifier effectively multiplies the detectorsignal with the sinusoidal reference signal at a frequency nf and then takes a time-average ofthis signal. This means that the nf lock-in output signal is given by equation (26).

−10 −5 0 5 10

Detuning frequency, νd

WMSsign

al(a.u.)

S2S4S6

−10 −5 0 5 10

Detuning frequency, νd

WMSsign

al(a.u.)

S1S3S5

Figure 5: Lock-in amplifier output signal, Sn, under pressure broadened conditions (Lorentzianlineshape) as a function of detuning frequency for detection at different harmonics n.

4 Experimental setup

A schematical layout of the experimental setup is shown in Figure 6. The source of light is adistributed feedback laser (DFB) laser in a 14pin butterfly mount, with a center wavelength of1570.3 nm (see attached specifications). The laser wavelength is controlled by a laser currentdriver and temperature controller and the output of the laser is fiber-coupled. The fiber isconnected to a collimating lens. The gas sample is pure CO2 at a pressure of 600 Torr in a20 cm long glass cell. An InGaAs detector with adjustable gain and bandwidth (see attachedspecs) is used for the detection of transmitted light. The laser wavelength can be tuned byconnecting a voltage signal (ramp) from a function generator to the input of the laser currentdriver. The modulation is provided by the lock-in amplifier (SR830), which produces a sinusoidalvoltage signal and also demodulates the detector signal at a harmonic of the reference frequency.The demodulated signal from the lock-in amplifier is sent to a data-acquisition card for signalprocessing and analysis.

10

cellcollimator detector

function

generator

laser

controller

ramp

modulation

lock-in amplifier

2f

MatLab

laser

Figure 6: Experimental layout for wavelength modulation spectroscopy.

5 Tasks

5.1 Direct absorption spectroscopy

1. Set the laser current to 128 mA and the temperature to 6.5 kΩ. Make sure that thetemperature is regulated before supplying current to the laser! Align the laser beamthrough the gas cell and onto the detector. Observe the signal on the oscilloscope.

2. Connect the function generator to the input of the current driver and set the scan am-plitude to 1 Vpp and the frequency to 20 Hz. Scan the frequency of the laser and observehow the signal changes. The intensity incident on the detector changes with laser current.Can you see a CO2 transition?

3. Record a spectrum of the transition with the provided program.

4. How large is the on-resonance absorbance?

(α0L) = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5. Does it agree with your calculations?

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6. What is the signal-to-noise ratio?

SNR = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11

7. What is the detection limit?

(α0L)DASLOD = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

8. How far is this from the shot noise limit (a factor of...)? You need to know the powerincident on the detector and the detection bandwidth.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5.2 Wavelength modulation spectroscopy

1. After characterizing the performance of direct absorption spectroscopy, go over to wave-length modulation spectroscopy.

a) Connect the sine output of the lock-in amplifier to the external modulation input ofthe current driver. Set the frequency to 15 kHz.

b) Connect the output of the detector to the input of the lock-in amplifier.

c) Connect the output of the lock-in amplifier to the oscilloscope.

d) Set the lock-in amplifier for detection of the second harmonic. Set the integrationtime to 100 µs and the modulation amplitude to 30 mV.

e) Slowly increase the modulation amplitude and observe the signal on the oscilloscope.

2. Can you see any signal from CO2? Optimize the 2f WMS signal by adjusting the phase,gain and time constant settings.

3. Record the signal with the provided program.

4. What is the signal-to-noise ratio?

SNR = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5. What is the detection limit?

(α0L)WMSLOD = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6. How far is this from the shot noise limit (a factor of...)? Note that the detection bandwidthis different now! It is given by 1/(2πτ).

12

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5.3 Conclusions

1. How much better is WMS than DAS? Compare the detection limits.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2. Do you think it is worth it?

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

References

[1] W. Demtroder, Laser spectroscopy., Laser Spectroscopy, Springer-Verlag, 2008.

[2] Aleksandra Foltynowicz, Fiber-laser-based noise-immune cavity-enhanced optical heterodynemolecular spectrometry, Ph.D. thesis, Umea University, Department of Physics, 2009, p. 145.

[3] Pawel Kluczynski, Wavelength modulation spectrometry – a new description of its funda-mental principles and properties, Ph.D. thesis, Umea University, Department of Physics,2002, p. 47.

[4] Pawel Kluczynski, Jorgen Gustafsson, Asa M. Lindberg, and Ove Axner, Wavelength mod-ulation absorption spectrometry : an extensive scrutiny of the generation of signals, Spec-trochimica Acta Part B: Atomic Spectroscopy 56 (2001), no. 8, 1277–1354.

[5] Lemthong Lathdavong, Development of diode laser-based absorption and dispersion spectro-scopic techniques for sensitive and selective detection of gaseous species and temperature,Ph.D. thesis, Umea University, Department of Physics, 2011, p. 118.

[6] P.W. Milonni and J.H. Eberly, Lasers, Wiley Series in Pure and Applied Optics, Wiley,1988.

[7] L. S. Rothman, I. E. Gordon, A. Barbe, Benner, P. F. Bernath, M. Birk, V. Boudon, L. R.Brown, A. Campargue, J. P. Champion, K. Chance, L. H. Coudert, V. Dana, V. M. Devi,S. Fally, J. M. Flaud, R. R. Gamache, A. Goldman, D. Jacquemart, I. Kleiner, N. Lacome,W. J. Lafferty, J. Y. Mandin, S. T. Massie, S. N. Mikhailenko, C. E. Miller, N. Moazzen-Ahmadi, O. V. Naumenko, A. V. Nikitin, J. Orphal, V. I. Perevalov, A. Perrin, A. Predoi-Cross, C. P. Rinsland, M. Rotger, M. Simeckova, M. A. H. Smith, K. Sung, S. A. Tashkun,J. Tennyson, R. A. Toth, A. C. Vandaele, and J. Vander Auwera, The HITRAN 2008 molecu-lar spectroscopic database, Journal of Quantitative Spectroscopy and Radiative Transfer 110(2009), no. 9-10, 533–572.

13

[8] Jonas Westberg, Faraday modulation spectroscopy – theoretical description and experimen-tal realization for detection of nitric oxide, Ph.D. thesis, Umea University, Department ofPhysics, 2013, p. 148.

14

NEL Laser Diodes

FEATURES* Wavelength Range 1565 - 1625 nm, ITU-T grid wavelength* Fiber Output Power 10mW

ABSOLUTE MAXIMUM RATINGS(Tsub=25deg.C)

ELECTRICAL/OPTICAL CHARACTERISTICS(Tsub=25deg.C)

∆T=| Tcase-Tsub |

1 These products are subject to export regulations and restrictions set force by the Japanese Government.2

3 The characteristics which are not specified in the data sheet are not guaranteed.4 The characteristics under the different operation conditions from the ones specified in the data sheet are not guaranteed.

March2005

100 nA

20

Typ. Max. UnitsV

+1

* Data is not attached.

VPE

Cooling capacity* ∆TPE

Isolation* Is

CW,VDR=5V

Thermister resistance* R

Peltier current IPE

Peltier voltage

-0.5

Monitoring Current(PD) IR(E)

Tracking error ER

Dark current(PD) Ir(0)

Spectral linewidth* ∆νSide mode suppression ratio SMS

Fiber output power φe

Peak wavelength λp

Forward voltage VF

Threshold current I(TH)

NLK1L5EAAA1565-1625 nm DFB laser diode in a butterfly-type 14 pin package with thermo-electric cooler.Pigtail fiber is connectorized with an FC/PC connector.

Parameter Symbol Ratings UnitsLaser diode reverse voltage VR 2.0 VLaser diode forward current IF 225 mAOperating case temperature Tcase -5 to 70 deg.Storage temperature Tstg -40 to 85 deg.Photodiode reverse voltage VDR 10 VPhotodiode forward current IDF 10 mAPeltier current IP 1.4 A

Parameter Symbol Condition Min.IF=30mA 1.2 1.6

mACW,IF=80mA 10 mWCW 10

nmCW,φe=10mW 2 MHz

CW,φe=10mW -1 ITU-T

dB

dBCW,φe=10mW 0.1 mACW,φe=10mW 35

+0.5

1φe=10mW,Tcase=70deg. 45

IR(E)=constant

deg.A

Tcase=-5 to 70deg. 2 VTcase=-5 to 70deg.

NTT Electronics Corporation reserves the right to make changes in design,specification or relatedinformation at any time without prior notice.

Tsub=25deg.kΩ

30 dB

Tsub=25deg. 10

! WARNING

If you plan to use these products in equipment which could endanger lives in the event of a product failure,please consult an NEL engineer before usage. Improper application of these products may endanger life. Toavoid possible injury, make certain these products are used in a redundant configuration.

Chapter 6: Specifications

13058-D02 Page 8

Chapter 6 Specifications2

Performance Specifications

0 dB Setting 40 dB Setting

Gain3 (Hi-Z) 1.51 x 10

3 V/A ±2% Gain

1 (Hi-Z) 1.51 x 10

5 V/A ±2%

Gain1 (50 Ω) 0.75 x 10

3 V/A ±2% Gain

1 (50 Ω) 0.75 x 10

5 V/A ±2%

Bandwidth 17 MHz Bandwidth 320 kHz

Noise (RMS) 600 µV Noise (RMS) 300 µV

NEP (@ λλλλp) 6.0 x 10-11

W/√Hz NEP (@ λλλλp) 1.4 x 10-12

W/√Hz

Offset 5 mV (10 mV max) Offset 10 mV (20 mV max)

10 dB Setting 50 dB Setting

Gain1 (Hi-Z) 4.75 x 10

3 V/A ±2% Gain

1 (Hi-Z) 4.75 x 10

5 V/A ±2%

Gain1 (50 Ω) 2.38 x 10

3 V/A ±2% Gain

1 (50 Ω) 2.38 x 10

5 V/A ±2%

Bandwidth 8.5 MHz Bandwidth 90 kHz

Noise (RMS) 320 µV Noise (RMS) 475 µV

NEP (@ λλλλp) 1.0 x 10-11

W/√Hz NEP (@ λλλλp) 1.5 x 10-12

W/√Hz

Offset 6 mV (12 mV max) Offset 15 mV (40 mV max)

20 dB Setting 60 dB Setting

Gain1 (Hi-Z) 1.5 x 10

4 V/A ±2% Gain

1 (Hi-Z) 1.5 x 10

6 V/A ±5%

Gain1 (5 0Ω) 0.75 x 10

4 V/A ±2% Gain

1 (50 Ω) 0.75 x 10

6 V/A ±5%

Bandwidth 1.9 MHz Bandwidth 33 kHz

Noise (RMS) 310 µV Noise (RMS) 850 µV

NEP (@ λλλλp) 3.0 x 10-12

W/√Hz NEP (@ λλλλp) 1.5 x 10-12

W/√Hz

Offset 6 mV (15 mV max) Offset: 20 mV (75 mV max)

30 dB Setting 70 dB Setting

Gain1 (Hi-Z) 4.75 x 10

4 V/A ±2% Gain

1 (Hi-Z) 4.75 x 10

6 V/A ±5%

Gain1 (50 Ω) 2.38 x 10

4 V/A ±2% Gain

1 (50 Ω) 2.38 x 10

6 V/A ±5%

Bandwidth 775 kHz Bandwidth 12 kHz

Noise (RMS) 300 µV Noise (RMS) 1.5 mV

NEP (@ λλλλp) 1.25 x 10-12

W/√Hz NEP (@ λλλλp) 2.0 x 10-12

W/√Hz

Offset 8 mV (15 mV max) Offset 40 mV (200 mV max)

2 All measurements performed with a 50 Ω load unless stated otherwise.

3 The PDA10CS has a 50 Ω series terminator resistor (i.e. in series with amplifier output). This forms a

voltage divider with any load impedance (e.g. 50 Ω load divides signal in half).

InGaAs Biased Detector

Page 9 Rev A, March 28, 2012

Electrical Specifications

Detector InGaAs PIN

Active Area Ø1.0 mm (0.8 mm2)

Wavelength Range λ 800 to 1700 nm

Peak Wavelength λ p 1550 nm (Typ)

Peak Response ℜ( λ p) 0.95 A/W (Typ)

Amplifier GBP 600 MHz

Output Impedance 50 Ω

Max Ouput Current IOUT 100 mA

Load Impedance 50 Ω to Hi-Z

Gain Adjustment Range 0 dB to 70 dB

Gain Steps 8 x 10dB Steps

Output Voltage VOUT 0 to 5 V (50 Ω)

0 to 10 V (Hi-Z)

General

On/Off Switch Slide

Gain Switch 8 Position Rotary

Output BNC (DC Coupled)

Package Size 2.76" x 2.06" x 0.88"

(70.1 mm x 52.3 mm x 22.4 mm)

PD Surface Depth 0.16" (4.1 mm)

Weight, Detector Only 0.15 lbs

Accessories SM1T1 Coupler

SM1RR Retainer Ring

Operating Temp 0 to 40 °C

Storage Temp -55 to 125 °C

AC Power Supply AC – DC Converter

Input Power4 31 W

100 – 200 VAC (50 to 60Hz)

220 – 240 VAC (50 to 60 Hz)

4 Although the power supply is rated for 31 W the PDA10CS actual usage is <5 W over the full

operating range.

Chapter 6: Specifications

13058-D02 Page 10

6.1. Response Curve

800 1000 1200 1400 1600 18000.0

0.2

0.4

0.6

0.8

1.0

1.2PDA10CS

Res

pons

ivity

(A/W

)

Wavelength (nm)